Search found 413 matches

by permaximum
Sat Jun 30, 2018 3:53 pm
Forum: APBRmetrics
Topic: Open source historical RAPM data and code
Replies: 12
Views: 10904

Re: Open source historical RAPM data and code

I might be confused. Are you talking about his alpha=2900? If so that number sounds are right as alpha in sklearn is N/2 * lambda in R where N is your number of observations. If not then yeah my bad I thought you were talking about something else. Yes, I'm talking about that "2900" and the sequence...
by permaximum
Sat Jun 30, 2018 7:29 am
Forum: APBRmetrics
Topic: Open source historical RAPM data and code
Replies: 12
Views: 10904

Re: Open source historical RAPM data and code

After further investigation, your lambda (alpha in your code) values are too high. As a result of that you punish players with relative low possessions even more. You should change your sequence to come up with a better one. I would focus on the cross validation code there. I don't know how python ...
by permaximum
Sat Jun 30, 2018 12:24 am
Forum: APBRmetrics
Topic: Open source historical RAPM data and code
Replies: 12
Views: 10904

Re: Open source historical RAPM data and code

Nice job! Also very kind for you to share all this code. You are the first to share such code at this scale in this forum. So thank you so much for that. At first I must say I'm not a computer scientist, I'm no analyst, I'm not getting any money. I'm just a basketball fan (not even that nowadays) an...
by permaximum
Fri Jun 30, 2017 11:44 pm
Forum: APBRmetrics
Topic: GM adjusted plus - minus?
Replies: 6
Views: 4804

Re: GM adjusted plus - minus?

Almost impossible because of the lack of data. Any work on this would likely be statistically insignificant. Results would probably lie between the residuals.
by permaximum
Fri Jun 30, 2017 11:37 pm
Forum: APBRmetrics
Topic: BPM - RPM Blend
Replies: 15
Views: 13295

Re: BPM - RPM Blend

2/3 - 1/3 ratio is not ideal but should be close enough for predicting team wins of the next season.
by permaximum
Mon Jun 12, 2017 10:21 am
Forum: APBRmetrics
Topic: Hot Hand
Replies: 9
Views: 6544

Re: Hot Hand

If we had to call upon maths to prove everything and advance that way as the human race, our race would already be extinct. Even now you can see some completely conflicting findings on most of the new research areas and both sides claim they have statistical evidence for it. You just need to look fo...
by permaximum
Sun Jun 11, 2017 7:55 pm
Forum: APBRmetrics
Topic: Hot Hand
Replies: 9
Views: 6544

Re: Hot Hand

If hot hand is not proven by analytics it's only because analytics are not there yet. No one can convince me there's no such thing as hot hand with the usual very poor statistical arguments. I'm a former basketball player that played as an SG at school-level and I know the whole deal. If these so-ca...
by permaximum
Fri Jun 09, 2017 7:02 am
Forum: APBRmetrics
Topic: BPM - RPM Blend
Replies: 15
Views: 13295

Re: BPM - RPM Blend

I didn't mean to doubt it's objectiveness. Your reasoning sounds fair. I just wanted to know your exact purpose :) . I mean like predicting wins, predicting spread, individual player impact...
by permaximum
Thu Jun 08, 2017 9:07 pm
Forum: APBRmetrics
Topic: BPM - RPM Blend
Replies: 15
Views: 13295

Re: BPM - RPM Blend

Objective?
by permaximum
Sun Jun 04, 2017 10:09 pm
Forum: APBRmetrics
Topic: Why are possessions et al. estimated using crude formulae?
Replies: 4
Views: 3277

Re: Why are possessions et al. estimated using crude formula

I always use the first formula. In some cases, I used even more sophisticated formulas. I don't like "too much estimation" or "bias". I always try to minimize that factor in any scenerio.
by permaximum
Sat Jun 03, 2017 5:44 pm
Forum: APBRmetrics
Topic: 2017 Finals Predictions
Replies: 20
Views: 9785

Re: 2017 Finals Predictions

I would say 4-0 if I saw this thread before the first game and I'm saying 4-0 again after the first game. Last year I thought it was the common non-empirical 51%-49% ratio in favour of Cavs for me. Durant is better than people and +/- based metrics give credit for. Durant-Curry-Thompson-Green would ...
by permaximum
Sun May 28, 2017 11:40 am
Forum: APBRmetrics
Topic: Setting Up Excel File for RAPM
Replies: 33
Views: 25808

Re: Setting Up Excel File for RAPM

There have been people contacting me via pm or here in this thread or older threads about how to calculate RAPM and getting the data ready for regression in the past years. I assume there are still going to be some people interested in how to calculate RAPM whether they are forum members or they com...
by permaximum
Sun May 07, 2017 12:10 am
Forum: APBRmetrics
Topic: Poll: RPM's degree of efficacy in sorting players
Replies: 138
Views: 91176

Re: Poll: RPM's degree of efficacy in sorting players

Went through the calculation, and there was absolutely no change in the PA. Only the Bulls vs. Cavs game outcomes were affected by shrinking a notional, league-wide HCA from 0.630 and 0.615, and the only effect of this was to change the mix of unexpected outcomes but not the total. So, the figure o...
by permaximum
Sat May 06, 2017 6:33 pm
Forum: APBRmetrics
Topic: Poll: RPM's degree of efficacy in sorting players
Replies: 138
Views: 91176

Re: Poll: RPM's degree of efficacy in sorting players

To return the conversation to what might be considered an upper bound of predictive accuracy for whatever lagged metric on might choose, here's what I think is another representative estimate. I went through this past season's results, using contemporaneous SRS ratings (as taken from basketball-ref...
by permaximum
Sat May 06, 2017 5:38 pm
Forum: APBRmetrics
Topic: RAPM request thread
Replies: 155
Views: 173698

Re: RAPM request thread

Single https://docs.google.com/spreadsheets/d/1_JkGXDXtu3cSLzplkQo4mDYOMLi-Sh2YfIb9nKeMNZI/edit?usp=sharing Multi https://docs.google.com/spreadsheets/d/12rcLDCaJSh_b1hA3UyLG5-neayTBl_weqNVCGM0LheA/edit?usp=sharing (less weight to older seasons, 3 seasons total) How did you determine optimal weight...